Parameter-dependent effects of transcranial ultrasound stimulation on corticospinal excitability and inhibitory control
Xuewei Qin, Liyuan Ren, Xiong Jiao, Shanbao Tong, Qiang Hu, Junfeng Sun
Brain Stimulation 2026 · 10.1016/j.brs.2026.103199
Abstract
Transcranial ultrasound stimulation (TUS) with various parameters can modulate cortical excitability, but its neuromodulatory effects on excitability and behavior have not been consistently replicated across independent labs or compared within subjects. To examine the neuromodulatory effects of four TUS protocols against sham on motor cortical excitability and inhibitory control. Twenty-six healthy adults were enrolled to complete up to five sessions (four active TUS protocols and one sham); 24 completed all the five sessions, separated by at least one week. Motor evoked potentials (MEPs) were recorded at 20 min and 0 min before stimulation, and at 0, 20, and 40 min after stimulation. Stop-signal task (SST) performance was assessed before and after stimulation, and simultaneous EEG was recorded during the SST to extract event-related potentials (ERP). Linear mixed-effects models with false-discovery-rate correction were used for statistical analysis. The intermittent and continuous theta-burst TUS protocols (iTBUS/cTBUS), adapted from an animal protocol, produced excitatory and inhibitory MEP effects, respectively, that were directionally consistent with the effects commonly associated with iTBS- and cTBS-patterned TMS; however, this physiological similarity did not translate into the expected behavioral outcomes. The excitatory effects of two TUS protocols on motor cortex were replicated, yet MEPs and ERP changes were associated with divergent SST outcomes. ERP analysis showed that cTBUS significantly reduced P3 amplitudes, whereas other protocols produced no significant group-level ERP changes. The bidirectional iTBUS/cTBUS effects observed here provide preliminary human evidence partially consistent with observations from animal models, although mechanistic equivalence and full translational validity across species remain to be established. The excitatory effects observed with two previously described TUS protocols were directionally consistent with earlier reports. Nevertheless, TUS-induced MEP facilitation does not necessarily enhance inhibitory control in the SST, underscoring the need for task-specific multimodal frameworks to advance TUS as a precise neuromodulation technique.
Abstract via pubmed.
Exposures
Exposure 1: Four active TUS protocols (rTUS, tbTUS, cTBUS, iTBUS) vs sham over left M1
Target: primary motor cortex — “left primary motor cortex (M1), TMS-defined FDI hotspot”
Device: custom-built
| Waveform | pulsed, theta_burst | |
|---|---|---|
| Fundamental frequency (kHz) | 500 | ✓✓✓ |
| Pulse duration (ms) | 0.5, 20, 17, 17swept | ✓✓✓ |
| Pulse repetition frequency (Hz) | 100, 5, 30, 30swept | ✓✓✓ |
| Duty cycle (%) | 5, 10, 51, 51swept | ✓✓✓ |
| Sonication duration (s) | 900, 80, 20, 92swept | ✓✓✓ |
| Free-field pressure (kPa) | not reported | |
|---|---|---|
| Free-field Isppa (W/cm²) | 8.09, 2.68, 3.15, 3.15swept | ✓✓✓⚑ |
| Free-field Ispta (W/cm²) | not reported | |
| In-situ estimate | simulationmean or range across subjects | |
| In-situ pressure (kPa) | not reported | |
| In-situ Isppa (W/cm²) | 0.874, 2.64swept | ✓✓✓ |
| In-situ Ispta (W/cm²) | 0.0078, 0.35swept | ✓✓✓ |
Protocol A (rTUS): PRF 100 Hz, DC 5%, tone-burst duration 500 us, inter-stimulation interval 8 s, total duration 15 min. Protocol B (tbTUS): PRF 5 Hz, DC 10%, TBD 20 ms, ISI 0 s, total duration 80 s. Protocol C (cTBUS): four 17-ms pulses at 30 Hz intra-group PRF form a 200-ms group (51% pulse duty cycle within the group, 34% overall duty cycle over the group period), groups repeated continuously for 20 s. Protocol D (iTBUS): same four-pulse 200-ms group structure as Protocol C, organized into ten 2-s theta-burst trains separated by 8-s inter-train intervals, total protocol duration 92 s (6.8 s total ultrasound-on time). Sham used an identical setup with ultrasound output disabled, duration matched to Protocol C (20 s). Total acoustic dose (T x Isppa) was equalized across the four active protocols by slightly increasing Isppa for Protocols B-D.
Flags from extraction
n_subjects— 26 were enrolled and received at least one active TUS session, but only 24 completed all five sessions; two participants completed only a single protocol each.exposures[0].free_field.isppa_w_cm2— The Isppa values are described as 'scalp-incident' intensities measured via hydrophone after passing through the EEG cap-and-electrode assembly, not a raw open-water free-field measurement; classified as free_field as the closest available domain since it was measured with a calibrated hydrophone in a water tank.exposures[0].in_situ.reported_as— The simulated in-situ Isppa/Ispta values are reported as a range across the four protocols (parameter sweep), not a range or mean across subjects; 'mean_or_range_across_subjects' was used as the closest available vocabulary option.exposures[0].timing.waveform— Protocol A (rTUS) is a standard pulsed protocol; Protocols B (tbTUS), C (cTBUS) and D (iTBUS) are explicitly called theta-burst by the paper, though the authors caution their theta-burst timing is not equivalent to standard TMS theta-burst stimulation.randomised— Randomisation refers to counterbalanced order of the five protocols across sessions (balanced Latin square), not allocation to independent treatment/control groups.